B1: First two chapters of N. Deo (the simulation book)
B2: Chapter 15 of CS229 (Reinforcement learning).
B3: Reinforcement Learning: An Introduction (Book by Andrew Barto and Richard S. Sutton)
Examples of simulation, e.g., 2D/3D motion of an object, pursuit problem, Inventory problem and queuing problem---B1
The RL framework (Agent, Environment, action, state, reward), MDP formulation, Value function, Policy function, Bellman Equation, Policy iteration to determine optimal policy--B2
Multi-arm bandit (MAB): Definition of n-armed bandit problem, Incremental Implementation, Nonstationary problem, Optimistic Initial values, Upper confidence bound (UCB)--- My notes, B3
Temporal difference (TD) learning: Monte Carlo method, TD(0), Idea of Model free learning, On-policy and Off-policy learning, Examples--- My notes, B3
On-policy learning: State Action Reward State Action (SARSA), Gridworld problem--My notes, B3
Off-policy learning: Q-learning, Cliff walking---My notes, B3
Actor-critic method: http://incompleteideas.net/book/ebook/node66.html
Introduction to Deep Learning (Lectures 30-34): https://www.youtube.com/watch?v=zGQjh_JQZ7A&list=PLJ5C_6qdAvBGaabKHmVbtryZW9KpICiHC&index=31
Deep Q Learning (DQN): My notes.